Cisco Introducing Cisco DNA Spaces

again thank you very much for coming my name is Darrell slidin I'm a at Cisco I'm a product manager been there for a long time right now they've been while the product manager in our location space spaces and our location area of our wireless network and so what we've really been doing is focusing on a pretty dramatic change that's just happened recently with last come on I talk to tech field day we were just about to announce what we're doing now and now we've done which is sort of rebranding and creating a new concept called our location as a platform which is now called Cisco DNA spaces and so with the acquisition last year of a company called July systems that provided us a network base cloud we were really we were able to take a lot of the individual components that we previously had so we had lots of great location mathematical engines right that ran on pram on a server we had lots of great cloud services we had lots of great api's but was pretty just just disjointed what we really focused on was with DNA spaces trying to put that all together and so why are we really kind of do that because it's been this sort of dramatic twist in a lot of industries right digitization really understanding what's going on with your network what's going on in your physical space has been the thing that unlocked a huge amounts of value right if every retailer could understand as well in the physical spaces as well as Amazon does in your digital space they would have a lot better chance of becoming you know the people who can compete against Amazon because amazon knows how long did I spend in baby diapers how long did I spend an outdoor furniture but when you're in the store they don't have that visualization or information about you at this conference for Cisco live we need to know are you a person who spends more time at dev net or the hub we offer Japanese food in Italian what's the difference for people hanging out where do people like to hang out as we know more about this conference through digitization we can offer better conferences so this sort of digitization of indoor space applies lots and lots of industries but one of the important things is it really has to happen at scale and so a good finding is that there's always this sort of digital indoor digital blind spot right whether it's uber out knowing your GPS location where do Pete most people gather to pick up cars or or other information that is available and traditionally from companies like that provided GPS mapping so they really understood what was going on outdoors that indoor space at scale hasn't been available and so what you find out is this that there's this limited visibility in your indoor space so when we ran these conferences five years ago we kind of you know got feel know what what area you were was in most more than others but in order to really know what's going on we can start to take advantage of this giant data feed that we're getting from the network so and that's really about can your network deliver more than just connectivity it's this giant feed that we get from the network so that that indoor visibility when you decide to you know use your passport your phone as a passport of course fully compliant meaning they decide don't want to do it I say nope to those terms and conditions are not included but if I do there's great applications and some applications that you might have seen at the Cisco live out here are the mapping application and find my friends you can do a find my friends application based on I'm going to share my location with a small group of people and they'll all be able to know just during this conference where I am you know it's a great way of being able to take advantage of the network adding something more than just connectivity on top of your network and so what we've done with DNA spaces is package this into sort of three levels see act and extend and the idea is there's always been lots of regular IT use cases and those are all part of C and then we've got some additional API base use cases that are an active extent and so the idea with C is and we'll show a demonstration of all this stuff is that you're able to unlock the value that you normally had with IT so I T always used to know where were people when they complained about Wi-Fi where are my rogue devices who is broadcasting somebody who is broadcasting Cisco a lot of 2019 but that's not really Cisco live 2019 all those regular use cases are part of our seats as well as using the basic API to understand what we're devices our act is really the area that we focus most on which has been adding an enterprise level kind of if this then that level on top of it so apt we can create these rules that can be triggered in the hundreds of thousands and millions because based on our cloud that can say if this device leaves this area if advice enters an area if a temperature alarm is triggered if a person comes back four or five times we can create these very sophisticated rules and that's what our act engine is all about is taking not only this network feed that a lot of the programs are saying yeah I know I could do this myself but I wish I could hand this off to somebody else to be able to do this so we've provided in this one purchase against your your ear API against your APs we just buy this one service and now I have a whole set of engines that I can use and then finally what we're trying to focus on with the extend is opening up the concept of a location App Store I think it's never existed before the industry kind of like and we're magnet there we're basing a lot on the some of the success that Salesforce has had and we've actually had some really good interesting meetings with them to be able to say Salesforce took a CRM data that was always destroyed lots of different ways of doing it and now I've got a centralized customer relationship information and I built applications that are on top of it whether it be marketing applications whether it be applications about how I can do service to my customers taking us a regular set of standardized set of data and having third parties add applications on it has really been successful in the CRM world we're trying to bring some of that six that capabilities to the location world so imagine that once I have a very standardized set of location data I can start to provide this to multiple different vendors whether it be maize map to be able to say I want to do an indoor wayfinding whether it be local measure to says I want to do a sentiment analysis whether it be rough in Editha says I want to be able to do a which rooms are most utilized workspace analysis there are lots of different verticals that need good location data and right now it's always been yep got a row right back down to the individual stack to write that and then the other company would go right back down to this right that and so what we're trying to do is each of those companies would come across the DNA spaces so you take your location data from your on-premises pointed at a trusted relationship with Cisco that you would have and now you can be able to say hey Cisco I I know you have a trust relationship with roof initi or a local measure or Mays math can you broker this location data so that when I say it stops it stops if it's GDR compliant if I say pull it MAC address out of the database it's pulled out of all of your apps at a single place if I want to be able to only give this person access for a short period of time I can start to do that so I have a single place where I can create application rules that apply across multiple applications that's really a focus on our our location App Store sort of extend level you know really we see a lot of business value in each of these areas right so in C we're just gonna improve the user experience drive efficiencies business decision making and business impact meeting the more I know about how many people have joined Wi-Fi where they have in a consistent manner I can compare it across other networks so the idea in web logs is there's great sort of well-defined metrics as how do i define a visit to a website now we can take a lot of that information against your physical space and have these well-defined metrics that everyone will use sort of these same metrics you can start to be able to compare them and we can provide things like an industry average or an organizational average and you can start to compare your sites across them in act it's really that there's a lot of business value in the the idea of once I can interact with somebody I can influence behavior right if I can say hey this is the third time you joined Wi-Fi this week as a proxy for this is the third time you've been here this week hey I want to be able to incent you to come back the fourth time by saying hey if you come back tomorrow I'm gonna give you a coupon for 20% off being able to do that on a one-off basis there's lots of different ways of doing that but being able to do that at scale and we've been able to prove this because we already have half a million ApS connected to this platform being able to do that at scale for like three thousand different locations that's really the value that you start to see with the X basis Act you can start to do these if this than that rules at very large scales that influence behavior and do things like influence and increase customer engagement and satisfaction because I can understand you know what you're doing yeah whether or not there's been a problem connecting any sort of sort of issues that you've previously have had and then finally the other part of the extent is also into a lot of the enterprise integrations so Enterprise integrations have great set of integrations into hey this customer is connecting to this website this often right and so that they have a really deep understanding about your customer experience on the web but they don't have a really great experience understanding of your experience as you physically go through a theme park or a hospital or any other areas right and so the idea here is we if we can integrate with enterprise applications we can start whether it be like an Adobe sales full Adobe application or Salesforce if we can start to put really good location data into your CRM data we can really we can better understand those customers so any questions of our serve are see active extend as sort of the theory I just wanted to start this presentation off with that and then we'll go into the dips of the individual applications and even some of the new hardware that we're going to be deploying to make this possible questions oh well well well keep it going we'll see what whether whether I can get some questions out of it because we're asked the most questions we'll get some new hardware that will be showing off but so I will be giving stuff away both to the people on camera somebody on camera we'll get them as well as somebody here in the audience a couple of people here so again DNA spaces see a can extend we're really focusing on scale when we've when we talk about scale this is something that we've learned from the Meraki runs our WebEx brother ends July systems at Meraki have talked about being able to do this at a very large scale is something that we've brought in house with our acquisition of July and being able to integrate and trigger thousands of thousands or millions of individual interactions with these customers and then of course having this as 24/7 support so if I have decided I put my location engine and I've moved that on-premise this location engine into the now my cloud is constantly calculating where locations are if there's ever a problem with that so I've had my Prem I on premises a pees like there's an AP right there let's go live we're sending this data off to the cloud now the cloud can report back he wait a second there's an AP I haven't heard from for awhile there's something wrong with that ap so we can start to have a better 24/7 monitoring because the network application that we're sending data to can also be aware that it should be receiving this data customer can hosted in their own data center if they have regional let's say loss or something like that yeah yeah good question and so this one we've leaned in the public cloud so this is cisco is offering this in clouds in Europe and the US you can decide which cloud you want to host it on in the Amsterdam cloud or the u.s. cloud but it is DN a spaces is only offered as a cloud service and software as a service so it's subscription only cloud only and you pick your date you pick your sort of world region that you're going to deploy this in and we've done this really to focus on the speed of deployment right the installation script and on-premises solutions are great and we've done that for a small component so we've left the location engine would be the last thing that will still exist on premises where you can make the choice to say no I don't want to touch the cloud at all I'll just leave a great location engine where we run some mathematical models that we've developed over the last 10 years as to the way of estimating the location of device based on our versus re signals that model can be run either in the cloud at infinite scale or on your premises but all of the additional applications are all cloud only not as you mentioned cloud only there are a few countries in Europe where you're not supposed to export personal data and this is personal data into the cloud that is hosted anywhere outside of that ya zone so where do you have your cloud locations so our cloud vacations are an Amsterdam and then us and so for those countries that say no I don't want to do it we offer an on-premises box and all of the applications that are on top of it they just will not be able to play for that particular country which will allow you know individuals in that country to write their own applications and things like that but we offer a solution that is European hosted and a cloud service that offers reliably enough for Switzerland right not good and not gonna for Russia right and so we make a business decision as to whether or not to move the data set and put a data center in Russia to do these applications but you know we're trying to be able to do this as fast as possible and we really lean into by doing this by doing this on the cloud just a quick one here on when you design this yeah I know Cisco's the US company but we are here to school live Europe yeah do you have already this something like this on the focus when you say okay if we design it in a cloud only fashion maybe we will not be able to deliver it for example to Russia to Switzerland or is this something that came up later in the development process no it really came up very early because we were gdpr compliant but that was one of the first things that we really focused on how can we make this compliant so we rattled you know we had our lawyers go through a lot a lot of different red legislation as to what types of compromises we could make and you know there were sort of positions of yeah if we put a mini data center in every applicant in every country what could we do if we made this as an on-premise application where you put in your own cloud what he can do and we we spent a lot of time weighing the cost versus a speed of that and the compromise was a European data center that would apply to probably 78 to 80 percent of the countries and a u.s. Dennett data center and in Asia pact data center we've also got the only validated solution for this in China so it was actually really hard we spent a lot of time getting a Chinese license which is really really hard and a Chinese data center to be able to do a rules engine for digitizing things in China so yeah we did we did spend a lot of time doing it was but we set to make some compromises on this uh Amsterdam us Indian China yes well a PAP region we have some P customers who go out at us some customers that want to go out of China based on the individual customer decisions some of them are its I yeah yeah it's not opting not to play yeah yeah there's where do i crusty or do i wanna some of them actually we have some customers that like the amsterdam data center regardless of it because latency is not that important in some of these applications if ice if i send you a notification of a text hey welcome back 250 milliseconds later because if that's the latency to get to Europe doesn't really change the experience right and that's one of the things that we're focusing on if you do need hey we've got two forklifts that are gonna collide into each other we don't do that in a cloud we say no no that's an on-premise this location engine that's something that you're gonna do separately but if you want to be able to do to set a alarm that a temperature has gone over 70 degrees you know moving that to an Amsterdam data center with 250 milliseconds latency has not been bad so one of the other things that we really that we focused on this I said really again I want to strangle myself basically really again has been a single plane of glass so that we've seen that there's lots and lots of different cloud services so we are I've been focused on crying to create a single place that you log on at DNA spaces dot Cisco calm or DNA spaces on i/o and then you can be able to get an overall view of what's going on in your network and see into these different types of applications whether it be captive portal engagement rules operational insights would be able to get the view of all of them across the single area so overall they the way that the this thing is being structured and then what we're sort of denouncing here at Cisco live and being able to talk about is this new DNA spaces unified cloud unified location cloud this location cloud is based on that you know we can connect to any type of network infrastructure including Meraki which was our Brethren's because the the company that we working a lot with as part of the acquisition which is July was one of what their luck was that one of the largest integrators on that so they've used that extensive API so we've got a really great experience with integrating and pulling data whether it be the location data or the captive portals off of Meraki and we can pull off old wireless Cisco existing networks so we built a cloud connector so even if you've got wireless back down the wireless LAN controller 8.0 with some of those de 1142 who ATP's out there that's you know built like tanks and lasted way longer than they should have you can still connect even that to DNA spaces right because we've built a effectively a translator so the translator says I'm going to speak on this leg the old n MSP protocol of report one six one one three that was not protoboys it's it's a TLV encrypted language that only Cisco knew and then on this other leg I'm gonna incorporate that into HTTP over protobufs it so that you know we sort of translate from the old world to a cloud enable world and that's what's called our cloud connector and that allows us to enact all the way back to our oldest controllers and then our newest controller some of the catalysts 9800 they will call it what I'll call natively speak cloud meaning they've got this great HTTP engine that can create packets and an HTTP 2.0 compliant method it could have multiple streams with TCP och when I'm sending packets when I'm sending some data out I can send it simultaneously out on five different channels wait for those acknowledgments to have a nice highly scalable solution that is native in the 9800 s in iOS controller 16 dot 10 and 16 11 so that's sort of the new version of there are catalyst engines that have a very high speed a highly capable native web connection capability and so for all the ones in the middle we've got rocky connectors catalyst connectors and translators that get us there so I guess deploying this is no big deal because most of the stuff is cloud-based yes was the sales model in particular how are these traditional system integration partners involved in this yeah so this the traditional system integration part we'll be able to pull all of the data that they need off of two different types of api's one is again this firehose API and so this firehose API is something that we're talking about now we're introducing it to continued customers and it is very similar to you know some of the the concepts that were previously heard about which is a atomic api meaning you can ask it to restart hey I missed your last three hours worth of data can you resend all of that so that's one of the capabilities that's being added as well as they can also do direct integrations with the components of the cloud so they have that sort of the choice either the partner API where they get this restart capability which is very effective or they can do the traditional I'm going to pull information from each of your individual components and I guess this is sold as sort of a subscription lock software as a service three five seven year licenses on a per ap per year basis the the big thing that we are doing there is you get all of this stuff so when we talk about this you'll get every single one of the capabilities inside of act when you buy this so whether you need it or not whether you're ready to consume it or not because maybe a better way I will charge it for but the the idea is we didn't want to be focused on these layering sort of yeah it's two bucks to do this and three dollars to do that and four dollars to do this you know we Rock to be very successful at you saying yeah we're not going to start slicing every single little feature into additional cost we understood what works well as you offer a package of features that become very compelling and so we offered the package really at two levels which is the sea level which is you know very minimal it's actually included in any of the new modern licenses that you buy things like DNA premiere includes that level and then the act level which is price per ap per year basis that allows you to to add all of this functionality with a single SKU and if you add additional features in the future they'll all be an act we are not gonna be like to say as I said today it's 2019 January that's what I'm pretty sure I'm pretty confident we can talk about though that's nice know adding a bunch of layered licensing on top we're not gonna yeah Pro that proact from super there's three months so a couple of the ones that are coming in this in this act level that we haven't completed are things like our ble manager and our ble manager is what I want to talk about here today one of the examples of something new and so this is what we're introducing this is a new USB dongle from Cisco this goes into your twenty eight hundreds or thirty eight hundreds and adds an a ble radio so something that we've been looking for for a while has been adding has been all of our new ApS anything since 2017 or forty eight hundreds or eighteen fifteen s they have native ble radios muraki's been having native a radius for several years now the thirty eight and twenty eighths which are a loss which our majority of a lot of networks didn't have a PLA radio this is the ability to add that ble radio to that so this ble radio is both the transmit and receive radio so it can transmit out a beacon buh-buh-buh-buh pop up that you can do all kinds of great new custom applications that can say oh I've detected this beacon you know send a welcome message or help me with my indoor wayfinding or this can be a receive radio so any asset tags so if I have an asset tag I can be able to detect this and be able to place the location of this and so these are all cloud managed ble radios that will be attached to your thirty eight and twenty eight hundreds would anybody like one of these okay does anyone have any twenty eight hundred or thirty eight hundreds no no twenty eight to thirty eight well if anyone has any 28 438 these are or eighteen fifteen or eighteen tens these are the dongle that you need to to enable that to this new functionality inside of your network yeah you'll update to the latest software and and you know what this certainly be included as part of that dumb that update so once you put this new dongle into a twenty hundred or a thirty eight hundred now you've taken that existing AP Network can be able to add this new radio function out nobody wants one hopefully some off the list so we were talking a little bit about some of the capabilities that you'll start to see with with these ideas there's a key use use cases that we think applies and you know we've posted these slides to make them available we think that across almost any industry better understanding what people and what things are doing in that physical space will make your net will make your business more efficient right whether it be education so we've been working with companies like degree analytics or involve EO companies like sensible business Sciences and they do things like hey this student has been on class Monday Tuesday Wednesday Thursday all of a sudden he's been missing class four or five days in a row they build machine learning models that say if a person who is previously always in class starts missing class what's the projected likely hint of this person succeeding or failing in class and so it's it's very interesting models that they can start to take advantage of because students almost always associate to the wireless they can start to do a tendency if it becomes mandatory in some places where you can say if you bring your device to the class we could Auto create attendance lists and so these things have been very successful in the US universities especially the US universities a little bit different than then it then the end the ones in Europe they are a lot of them are for-profit so they really want students to pass right they really don't want you to drop out so they use methods of being able to how the counselor interject if you've missed class for a couple of weeks to say hey are you sure you still want to be in this right so it's it's only one thing all it is is one signal to the counselor to say hey maybe you should reach out to the student he's stopped coming to class right there might be a good reason to it I think the success ratio the student not coming to class in a lot of the in a lot of the US university has a big impact as to whether or not they they are graduating so think of these four-year undergraduates you know classes that are not the highest level academics is not so much problem at at a Stanford or MIT but some of the mid tier universities that if a student stops coming to class if they got in if they got somebody who reached out and contacted them said why and help them get back to class there's a much higher chance of them graduating and I see people carrying backpacks of mobile phones into the classroom yeah whatever they can do to help them better learn in those classrooms and some of the other places that have done really successful by digitizing their spaces have been you know University of British Columbia uses this for an interesting use case where they decide when to turn on the HVAC system when to turn on heating and air conditioning because they you know some buildings if they follow the standard model that will warm the building at 7 a.m. or quote will well cool it off at 7 p.m. there's a lot of buildings that either start earlier and earlier in the day they are pretty much empty by 5:00 and if they were able to turn the air conditioning off one hour earlier they able to save a large amount of money and so use the the Wi-Fi signature of devices associated in the building as a very good proxy to determine the occupancy which then led to a reduction in the amount of air conditioning it's been required there's a lot of interest in your building yomo it's gonna start to get cold anymore but the building will save a million the building is spelled we probably spending ten thousand dollars to warm you yeah it's there for the environment you shouldn't be on it so but there are a lot of interesting use cases like that into it here in Europe has done a great job helping to view footfall traffic and be able to understand things like sporting events whether the person who purchased the ticket really King right did they are they selling your season tickets are they actually coming a lot of to make the changes of how I market to work sicles multiplexing so they can sell to many people now right so even if I used to it says you're sowing to the season ticket-holder start a twenty thousand spaces they can sell now maybe mininum guys so maybe just a comment it's not really related to technology so I'll try to keep it short I think that as a technology provider what you're doing is great but I think as a society we may run into a problem we are a two-speed society people which have electronic devices and which are digitized and which can benefit from all of the things and other categories of people we're maybe not so fortunate in life and we're all left behind you know so maybe that's a that's a gap or a divide that we need to kind of figure out go to kind of let's say compress already lacking at the bridge oh reading the markets yeah but that I think that that's a really astute comment truck Robbins was in Davos this week really talking about how you know still a third of the world isn't digitized they live on $2 a day how are we going to be able to to help that and I think that's going to be it may not make sense no because these things are emerging or we're just starting business we go into smart cities and I think that's some of the messaging that you've been pushing in the past years with IOT and all I think we may really reach that point where we really need to something about it yeah I I absolutely agree that that digital design what do we do for that digitization of the rest of population is it you know everything from you know subsidized smart phones as I as a human right right like baby I'll talk about the Internet's a human right or is it going to be something where only a certain amount of things can be done in a digital fashion it's a lot of great great thoughts on that you know going that we were talking about that about 24/7 monitoring because this is a cloud service where offer were able to offer a very extensive monitoring of the system making sure that the system's always up and running we're able to be able to have a set of administrators that are on then always watching the network and make sure it receives the data it's required and there's really four things that we wanted to focus on with DNA spaces one is speed of activation so we've actually created a guide that is at DNA spaces cisco.com that in four steps you can get any network connected like we were talking about how easy it is to connect the network whether it's a Meraki or a catalyst network that you know that speed and ease of acquisition of activation is something that's historically not been Cisco strong point right being able to get a controller updated or being able to get to a new set of capabilities hasn't been something that's easy and so what we what we have been focusing on with the isis is in four steps you can get any type of network connected so that's one of our main components the second one is to create this industry-standard data right so that as we receive location data meaning this device has been here seen seen seen by a probe then it's no longer it's no longer seen how do we define that we define that visit as if you seen four one you know if you only get to one probe packet is that defined as a visit does it have to be a certain number of probe packets so we've defined sort of a standard metric that a device is seen once he's gone for three hours from a network that's when we close the visit if it doesn't come back within three hours we can started a new visit we define these sort of standard metrics that we will apply across a bunch of different industries so that you can start to do comparisons of your location data so really focusing on reliability accuracy data normalization so whether it be Meraki network or a cisco network we can receive the location updates and normalize all the data to be able to say this is how we define a visit and then create these industry standard metrics so you can start to compare hey what's the average frequency of a person coming back to my network versus a person coming back to a competitor's network in the same industry and all this is done with with data kept compliancy privacy and compliancy meaning if i decide to opt out and say you know I don't want to connect it either on my terms and conditions on my connect or any other method once I opt out I'm not a concluded in any of these data sets the third is a very standard way of doing these behavior metrics how many of your locations performing against the key metric how are your locations doing relative to each other which metrics are relevant as to frequency of real visits where turning visits all of this data whether what's particularly relevant and then if I did some big event let's say I added signage or if I changed a routing I can do an impact analysis how were my metrics before that I do something how are my metrics after that so I can start to understand very much like you do a/b testing on a web I could do a/b testing in my physical space and be able to understand when I change my physical space is that um you know are there changes in people's behaviors before and after that and most of these things have always been done in a sort of gut feel basis right I kind of yeah the sign looks better over there I think people are going to be more appreciative of it right or yeah I think let's change the flow of this of this part of the store and it's been just sort of gut feels as to whether it's working what we hope to do with this is be able to say hey you can use some type of standard metric to count the way people have either spent time or repeat visitors and use that to help you compare items before and after and fourth is really that the big thing is our data integration into enterprise-wide systems whether it be you know loyalty systems that can say you know there's a great set of loyalty systems that you know every big enterprise has that has a crm number that says hey are you a hotel loyalty member or are you a little team member at a theme park or you know even the grocery stores but your physical and when you're you're you're also member often understand what you do on the web if you log in with your loyalty number but they don't necessarily understand where you go in a store they understand what you do that to check out because they have what you bought but they don't understand where I actually went during during a particular store so if you can start to integrate your in-store experience with your CRM data I can better understand oh this is a customer that goes to four different Nike stores before buying something this is a customer that spends a lot of time browsing this particular area in the store and so those are the four key components that I wanted that that are pardoned about about the a space is that ease of Opera ease of activation standard data business metrics and integration and let's take a look at some of these items I'm going to be sitting down and doing some of these demos all right first of all one of the dashboards that we've got here this is our sort of our overall dashboard this shows it shows you all of the information all the different networks that we currently have pointed to our DNA spaces cloud and so right now we have over 22,000 different locations where aps are placed and a total of over 500,000 ApS that are sending northbound data into our cloud and that means our cloud has processed over 140 billion updates so for machine learning a good data set starts at a billion we have a hundred and forty billion records so we're able to use this to do great sets of k factor analysis to say what's a what's an anomaly when we do a location detection because we can understand what things grouped together well and we're what things are outliers we can understand the different device behavior from an android device when it sends a probe probe probe and then stop sending a probe in a network we can understand how devices roam between different access points because every time there's a roaming event we get a new data location update so we have a really great set of data to be able to do some of our predictive as well as descriptive analysis on and so when we say we use our machine learning techniques to help you define what a visit is this means we've gotten all these location updates and we could help us find anomalies in the data set a couple of the ones that we do automatically now are things like detect determination automatically of what MAC addresses are belong to employees most likely and which ones are shoppers by saying oh I'm employee is likely to be there you know eight hours a day expiry in a longer period of time shoppers are more likely to be there for shorter periods of time interstitially and so we can start to group people and make a guess that hey we think this is an employee that's using your guest Wi-Fi SSID because they're connected to your wireless 40 hours a week things like that so we have some great set of information that we can learn about this identifiable tonight yeah we'll we haven't been able to predict any difference in probe behavior for males and female but we do certainly understand if they have a captive portal where we capture that information that information is included in part of our metrics so our DNA faces dashboard has an initial digitalization statistics and tell you what are the number of access points that are connecting the types of updates you've actually received in your network and all of this information is used to feed the different types of applications you'll see whether it be different captive portals different engagement rules or location persona let's look into a couple of them one is for example is behavior metrics and what we can do with behavior metrics is be able to say oh what's the for a retail customer what are the average visit duration what's the average visit frequency and I can say hey this is for my entire network I know that most people visit 1.6 times a month but downtown Denver in Philadelphia are more like I have more than the than the historical average and these are my least attractive locations right so I can start to say oh maybe I need to do something different in New York because what the statistics are showing that these are the least valuable locations when it comes to visitor frequency and the same thing with visitor duration and so you'll start to see we get the the system does this machine learning and it usually gets 30 days worth of data before it has a very high confidence interval so we felt we shoot for an over 80% confidence interval in the individual data sets that has historically occurred after about 30 days with the data before then you can see your raw data but we'll say that it's not going to be as complete as when we have a full 30 days worth of data to be able to say what monthly visits right we need kind of a month worth of visit to make those yes right and so that's really what the first business insights will get you is yeah we can show you that there's less visits per month in York more visits in Miami well and that's where you'll have to say oh it's because it's sunny in Miami it's in the middle of a winter storm in you are right the idea of this data set can start to put into other smarter data sets to be able to make you know really guess guesses as to why these are why these are occurring and so that digitization of people which is a you know research first here all the information about what's going on and then we can create a whole series of rules and so in gate that rules are like engagement rules let's do this one here so engagement rules are BR the ability to be able to do things like if I have a visitor I want to welcome them that person back if there's a notify the associate that a high-value guest has arrived and so I can see how that if this then rule is created right so it's literally a GUI creation of a rule that says if person if a person is on Wi-Fi he's entered the San Francisco Union area I can put any particular area begins leg for opted end users or in this case a CRM tag has been created that this is a high profile loyalty customer and at during a particular time range I could send the text message to a phone number or it can be anything that says hey a high profile customer first name last name says please walk has just walked in please welcome them so this is an example of a rule that can be created to make your business processes easier but this rule is not just hey I can do this cute little test on one thing I could do this for 3,000 properties which is occurring now for the 400,000 guests so this solution of a very highly scalable type of rules into this what we've built here for people and these rules can also trigger things like trigger API so we can do a JSON post to a digital signage that we've done in as part of an RF that says you know a little bit of the Big Brother sort of factor but depends on some stores that you may want that is welcome Darrell back to a particular store right so as I joined Wi-Fi I can get an individualized welcome message we've seen people being able to think about at a restaurant to imagine when they start doing iPad based menus this is your individual menu here's what you ordered the last time you were here right so you could start to really think of how this personalization scan can begin to a to occur so this kind of thing I mean first name last name is then stored European or yeah once they've opted in they sign in and they'll opt in and they say yes I agree to these terms this Alton's yes yeah because I would never this is like I mean I can see the point of giving you more data points as a business to find out what people want right but me I would come in like and invisibly clear I'm not here yeah and some people do that on the web too right they can use you know a browser and and high cookies and not include any cross scripting and then they get a different experience whereas if you if you allow cookies on the web you'll get a different experiences and so it's very something like that yeah absolutely anyway Cisco Facebook but you can like just using the MAC address I I guess you can say this is a high loyalty customer no we yeah so you know so you could but the system is designed so it your select that there's these an opt-in user all we know about the user is that they join and we only are doing these engagements for people who have joined them the Wi-Fi you leave join the Wi-Fi and so you've at least done hike address um just to or to recognize up yeah for make a dress for yeah poison if I am changing very frequently if I am changing video often yeah let you login actually have a question from Twitter Nick looper wants to know that all of these ApS lack the hyper location angle of a rival array like the 3702 and the add-on for the forty 802 could they do less hyper ble location yes they can with this new a USB ble radio you can do soo standard location of the individual devices so it determines based on signal strength three different ApS here from different signals and it can provide a location with that few meters with about six to seven meter location accurate so would that eliminate the need for like a hyper location I do not need a hyper location to react use this USB dongle to detect the location of ble devices all right there you go Nick there's your answer just a quick one we have yen Europe and as you hear everybody is kind of in the mindset of data privacy and so if you present this to the US I guess even in retail and so on the people are much more open with engaging to that or how is it think about what we found it in US and there's three places that are really interesting is hotels people in hotels want to be known what they want to be you want to use your guest Wi-Fi you want to be recognized you want to be recognized because it's almost like a your home your home away from home so in hotels we found a lot of people are opting in sports teams right so a sports teams I want to join that Wi-Fi because I'm a rabid fan I want to know that I'm a rabid fan people really like to say hey welcome back as a fan integrations and so we found that those experiences sports teams and hotels are kind of consistent whether it's Europe or the u.s. that they're about the same retail is one that's a little bit different sometimes retailers are are more proactive in the u.s. the Nike stores in the US are much more active about hey welcome back to this particular Nike store where that might be less used in Europe I think retail is different but hospitality and sports is actually more in Europe that is maybe people forget that about that privacy when they're when they're really routing for their team they want to do it they might fossile own us across the goal is everybody they all want interactions they want to be hey anyone who's who's joined the Wi-Fi gets a free whatever or coupon if Barcelona scores a goal right they like that type of interactions and so when you view these rules you get to have a lot of data about these individual rules that happen and then the other thing is how did that rule actually trigger right in this case we can say oh this rule triggered you know 700,000 times with these unique users how they're triggering it how that rule triggers by the time of day and how am I notified so when this rule triggered it was it a text message sent an email a push or an API right and so this is the idea of rules engines about people and then we also have same rules engines about things right whether it be USB tags whether it be laptops we have the same sort of rules engine so operational insights is the ability to be able to understand what's going on in my physical space I have lots of things in my physical space whether it be IOT temperature monitors or I've a lot of iPads and laptops that are used for public usage I want to be able to know where they are and what's going on with them I have the same very high scalable rules engine so in this rules engine I've got a similar sort of condition that says Darrell has left the Cisco live world of solution for my particular asset if the location has exited from the Cisco live immediately send me a text message that I've left right and so create these types of rules and these rules can apply against to the individual people or the thing and so that that applies across all of them and so I think that's really the messaging about the digitization of people and spaces di spaces doesn't set a very high scale all the information is then sort of exported via either api's and the third-party systems as part of our location app store or you can build applications and sort of experiences inside of the product right we've seen both kind of cases really important that some people want everything to be API driven but some people are like yeah I want to be able to build some stuff natively in the product and so when we say natively in the product is I can have an end-to-end system that says you know if a temperature of a refrigerator goes above this level send a text message to Steve right and then BOOM got an end-to-end sort of system that that that works and provides them at the definitive value to the customer all right that's really what I wanted to talk about with DNA spaces and so you know our takeaway was that maybe we'll have some of the people at home who asked a great question it has a 2800 we'd like to get a these new USB these ones are you know they're not they don't even have the FCC approval yet so they're they're a beta customer their beta a piece of equipment but we've added this as a way of digitizing things so not only your regular laptops ble tags as well as phones as well as ble devices all of this digitization is part of our overall DNA space how about that for her and the over my last call to action Oh because there were last minute I have this one call to action I was supposed to make sure I still talked about was what are you going to do to do this if anybody wants a free trial of DNA spaces they can right now go to DNA spaces on Cisco comm big click free trial is the middle of Avakian and they can connect their network for 90 days to the unique spaces and start to take advantage of all of these wonderful capabilities

Channel: Tech Field Day

Views: 6,685

Keywords:

Id: Poma9WSwv70

Length: 49min 16sec (2956 seconds)

Published: Mon Jan 28 2019